INDEX

Explanations

The neuron strongly activates on personal pronouns—especially first-person pronouns like “I” and “we.”

oai_token-act-pair · o4-mini Triggered by @jyhe0408

the pronoun "we" in personal narrative contexts.

oai_token-act-pair · claude-4-5-sonnet Triggered by @jyhe0408

personal subject pronouns in narrative text (e.g., first- and third-person references to people or things).

oai_token-act-pair · gpt-5 Triggered by @jyhe0408

New Auto-Interp

Configuration

google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 также

0.45

俩

0.44

 також

0.44

らは

0.43

ᅵ

0.43

inerary

0.42

 will

0.42

can

0.41

 بھی

0.41

brahim

0.41

POSITIVE LOGITS

the

0.63

its

0.54

our

0.52

any

0.50

it

0.46

 those

0.45

0.44

 some

0.43

an

0.41

 this

0.41

Activations Density 0.030%

The neuron strongly activates on personal pronouns—especially first-person pronouns like “I” and “we.”

the pronoun "we" in personal narrative contexts.

personal subject pronouns in narrative text (e.g., first- and third-person references to people or things).

No Comments

No Known Activations

The neuron strongly activates on personal pronouns—especially first-person pronouns like “I” and “we.”

the pronoun "we" in personal narrative contexts.

personal subject pronouns in narrative text (e.g., first- and third-person references to people or things).

No Comments

No Known Activations